Prerequisite/Recommended prerequisite for
participation in the module
Interaction Design, Mathematics for Multimedia Applications, Human
Senses and Perception
Content, progress and pedagogy of the
module
A crucial aspect of designing medialogy systems, tools or
applications is the need to evaluate the work experimentally. The
knowledge of how to properly design experiments to collect and
evaluate data is essential to answer many of the problems within
medialogy. Examples are testing which of two tracking algorithms is
the most efficient; how users perform with different kinds of
feedback; possible relationship between age and performance,
etc.
Learning objectives
Knowledge
- Must be able to understand the basic concepts
of probability: sample space of all possible events; combinatorics;
independent events; conditional probability; Bayes’ formula;
binomial distribution, etc.
- Must display knowledge about basic statistic
terminology and treatment of data: distributions (probability
density function, cumulative distribution function, quantile
function); measures of central tendency and variability; histogram;
central limit theorem, significance, power, type I and II errors,
etc.
- Must be able to understand advantages and
disadvantages with different types of designs and studies
(between-group and within-group designs; correlational studies;
blind/double blind, complete/incomplete and balanced/unbalanced
designs)
- Must be able to understand the difference
between common experimental designs, e.g., single sample
experiments, two sample experiments, and factorial/multifactorial
experiments
- Must understand the basic experimental design
principles of independence, randomization, replication, and
blocking and how these can be applied in experiments.
- Must be able to relate frequency distributions to the concept
of hypothesis testing (understanding)
- Must be able to understand possible ethical
concerns for a study
Skills
- Must be able to design an experiment to measure changes in a
dependent variable, identifying and efficiently controlling
relevant independent variables (application)
- Must be able to properly inform and instruct persons
participating in a study (application)
- Must be able to understand and select among
the most common methods for statistical analysis and assessment of
experimental data (e.g., t-test, analysis of variance, chi-square
tests, binomial test, correlation, and simple linear and logistic
regression)
- Must be able to understand the difference
between parametric and non-parametric analysis methods
- Must be able to understand different
measurement scales and discuss experiments in terms of reliability,
bias and sensitivity
- Must be able to discuss own data in terms of assumptions for
statistical testing (application)
- Must be able to use an existing statistical package to
analyse and present experimental results
- Must be able to discuss and represent empirical data in
different ways (describing text, numbers, formulas, graphs and
figures) and shift between these according to the needs of the
situation and context (application)
- Must be able to read, understand and implement experimental and
empirical work as described in relevant literature
(application)
Competences
- Students who complete this module will be able to
systematically design quantitative, scientific experiments, taking
into account relevant factors (application)
- Students who complete this module will be able to use a
statistical software package to analyse experimental data
(application)
- Students who complete this module will be able to document
their experimental results, and to understand experimental results
presented by others (application)
Type of instruction
Refer to the overview of instruction types listed in the start
of chapter 3. The types of instruction for this course are decided
in accordance with the Joint Programme Regulations and directions
are decided and given by the Study Board for Media Technology.
Notice: This elective course might not be offered if less than
10 students sign up
Exam
Exams